Messages in this thread | | | From | Nanning.Buitenhuis@consul ... | Date | Tue, 21 Apr 1998 13:17:59 +0200 | Subject | Re: [Offtopic] Unicode, kernels, Linux and NT |
| |
> > #define dev_name_str L"\\Device\\ise" > You're being unfair. The #define above already _is_ the Unicode string.
Alas, the above isn't true: If I remember the wide character definition of C(++) correctly, a 'long' string is zero extended. Now if you use 7-bit ascii, or even Latin-1, then the resulting code is functionaly unicode (but NOT the same). However, if you compile this on say, OS/390 OE or an AS/400, which use EBCDIC, you are in problems.
The problem is that the C(++) wchar_t is NOT unicode. On most ascii based platforms it usualy is (or the 31 bit ISO 10664), but on EBCDIC platforms it is something else. Totaly unimportant, until you want to write a truely transportable program.
Under C(++) there is no language supported manner way to write true unicode or ISO-10664 code (or UTF-8 for that matter).
I like to be proven wrong, but I did some extensive testing on this subject.
-- NaN (Nanning Buitenhuis) nanning@elvenkind.com
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu
| |